HCC 肿瘤免疫微环境的多维分析：预后意义及对免疫治疗的启示

The combination of immune checkpoint inhibitors (ICIs) and anti-angiogenic agents represents the standard first-line therapy for patients with unresectable hepatocellular carcinoma (HCC). However, most patients do not derive sustained clinical benefit. Emerging evidence has underscored a strong association between immunotherapy efficacy and the tumor immune microenvironment (TIME) in HCC. Accordingly, this study aimed to characterize the TIME and develop a potential approach for predicting patient survival.

We analyzed a cohort of 78 patients with unresectable HCC who received ICIs combined with anti-angiogenic therapy. Four multiplex immunohistochemistry (mIHC) panels were designed to comprehensively evaluate tumor-infiltrating immune cells (TIICs). Digital pathology was applied to raw imaging data to extract TIME features, including positive rates and spatial distributions of TIICs. Machine learning algorithms were then used to construct predictive models based on TIME-associated signatures (TIS).

High positive rates of CD103 + and CD103 + CD8 + T cells were associated with prolonged overall survival (OS). Conversely, a high positive rate of CD8 + PD-L1 + T cells correlated with shorter progression-free survival (PFS). A greater abundance of CD8 + and CD103 + T cells in close proximity to tumor cells was also associated with longer OS. Multivariate Cox models incorporating these cell populations demonstrated that a lower TIS was significantly associated with longer OS and PFS.

The composition and spatial distribution of immune cells critically shape the TIME in HCC and influence immunotherapy outcomes. TIS-based models show promise for predicting immunotherapy response in HCC patients, though they require further validation in larger prospective cohorts. These immune cell populations may serve as prognostic biomarkers and potential targets for personalized immunotherapy.